A subjective and objective integrated method for fraud detection in financial systems

Financial statement fraud (FSF) has cost market participants, including investors, creditors, pensioners, and employees, more than $500 billion during decades. Especially in recent years, with the worldwide use of financial systems in companies, governments and universities, fraud in financial systems can be in terms of computer, network, customer or even staff and all will remain keys in assessing financial system risk. Traditional methods such as auditing or statistics models used to detect fraud in FSF can't effectively select the intrinsic features in financial systems. This paper focuses on identity theft fraud in financial systems and proposes an integrated framework including subjective methods and objective models for fraud detection in financial systems. The subjective and objective integrated framework employs AHP and rough set (RS) to analyze the fraud scenarios, select the intrinsic features, detect the abnormities and alarm. The proposed framework used to detect identity theft fraud can be also used to detect and prevent other types of fraud in financial systems.